Drug Evaluation of Concurrent Pneumocystis carinii, Toxoplasma ...

ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, May 1998, p. 1068–1072 0066-4804/98/$04.0010 Copyright © 1998, American Society for Microbiology

Vol. 42, No. 5

Drug Evaluation of Concurrent Pneumocystis carinii, Toxoplasma gondii, and Mycobacterium avium Complex Infections in a Rat Model MONIQUE BRUN-PASCAUD,1* PREMAVATHY RAJAGOPALAN-LEVASSEUR,1 ` NE BERTRAND,1 LOUIS GARRY,1 FRANC ¸ OISE CHAU,1 GUYLE FRANCIS DEROUIN,2 AND PIERRE-MARIE GIRARD3 Institut National de la Sante´ et de la Recherche Me´dicale Unite´ 13 Ho ˆpital Bichat-Claude Bernard,1 Laboratoire de Parasitologie-Mycologie, Ho ˆpital St-Louis,2 and Service des Maladies Infectieuses, Ho ˆpital Rothschild,3 Paris, France Received 19 September 1997/Returned for modification 8 December 1997/Accepted 3 February 1998

We present a new experimental model for the simultaneous evaluation of the activities of drugs against Pneumocystis carinii, Toxoplasma gondii, and Mycobacterium avium complex infections. Rats latently infected with P. carinii were challenged with the MO-1 strain of M. avium and then immunosuppressed with corticosteroids for 7 weeks. At week 5 the RH strain of T. gondii was intraperitoneally injected. Organs were examined for the three pathogens after death or killing of the animals at week 7. Without treatment, rats challenged with T. gondii died with pulmonary P. carinii infection and disseminated T. gondii and M. avium infections. In order to assess the value of the model for evaluation of the activities of drugs, we administered by oral gavage for 7 weeks drugs or combinations of drugs selected for their individual efficacies against at least one pathogen. We found that clarithromycin with sulfamethoxazole, clarithromycin with atovaquone, roxithromycin with sulfamethoxazole or dapsone, and rifabutin with atovaquone were effective against the three infections, whereas PS-15 with dapsone and trimethoprim with sulfamethoxazole were active against Toxoplasma and Pneumocystis infections only. This triple-infection rat model offers a new tool for the simultaneous evaluation of the activities of drugs against three of the major opportunistic infections occurring in immunosuppressed individuals. Pneumocystis carinii, Toxoplasma gondii, and disseminated Mycobacterium avium complex infections are frequent opportunistic infections that occur in AIDS patients. Drugs with activities against these organisms are available, and the value of primary and secondary prophylactic drug regimens has been demonstrated (38). Current prophylactic strategies target both pneumocystosis and toxoplasmosis, when required (25). Combined prophylaxis for these two opportunistic infections has been achieved with two-drug regimens, i.e., trimethoprim-sulfamethoxazole (12, 27) or dapsone-pyrimethamine (20). The benefit of primary prophylaxis for M. avium complex infection with rifabutin, clarithromycin, or azithromycin has been demonstrated (29). Despite these advances, evaluation of new curative and prophylactic regimens for opportunistic infections is warranted to provide more effective and/or better-tolerated strategies. Animal models are major tools for this type of evaluation. Antipneumocystis drugs have mainly been evaluated with latently infected rats (17, 24, 39) or rats that are intratracheally inoculated (4) and then immunosuppressed. Antitoxoplasma drugs are evaluated with healthy mice (2, 15), and antimycobacterial agents are evaluated with immunodeficient beige or C57BL/6 mice (18, 30). Individually, these models provide evidence of the activities of drugs, but they are not well adapted in the setting of preventing multiple opportunistic infections in the same host. Therefore, our goal was to develop a rat model of multiple infections which takes into account the potential interactions of the organisms in the development of each infec-

tion and which permits the more rapid screening of drugs or combinations of drugs with extended activity. The use of a single host would also allow a simultaneous comparison of the respective efficacy of each drug against different pathogens under similar pharmacokinetic conditions. With this goal in mind, we have extended the rat model of concurrent infection with P. carinii and T. gondii (11) to a model of infection with three opportunistic pathogens by inducing active M. avium complex infection along with active P. carinii and T. gondii infections. After successfully establishing infections with the three pathogens and examining whether there is an interaction between them, we evaluated several drug combinations known to be effective in single- or double-infection models. (This work was presented in part at the 36th Interscience Conference on Antimicrobial Agents and Chemotherapy, New Orleans, La., 15 to 18 September 1996 [7a].) MATERIALS AND METHODS Triple-infection model. Male Wistar rats (weight, 180 to 200 g; Janvier Breeding Laboratories, Le Genest St. Isle, France) were used. This study was done in accordance with prevailing regulations regarding the care and use of laboratory animals in the countries of the European Community (Journal Officiel des Communaute´s Europe´ennes, 18 December 1986, report L358). The experimental protocol is summarized in Fig. 1. Induction of different infections. P. carinii infection was induced in latently infected rats by immunosuppression: 25 mg of cortisone acetate (Hydrocortisone; Hoechst-Roussel, Paris, France) was injected subcutaneously twice weekly, and the rats were given a low-protein (8%) diet (Usine Alimentation Rationelle, Villemoisson, France). M. avium complex strain MO-1, a clinical isolate from an AIDS patient, was prepared as described previously (30). At week zero rats were inoculated with 1.5 3 107 viable bacteria in a volume of 0.3 ml, which was injected intravenously into the jugular vein while the rats were under light ether anesthesia. T. gondii infection was induced as described previously (11). Briefly, after 5

* Corresponding author. Mailing address: INSERM U 13, Ho ˆpital Bichat-Claude Bernard, 46 rue Henri Huchard, 75877 Paris Cedex 18, France. Phone: 33 1 40 25 86 05. Fax: 33 1 40 25 86 02. E-mail: u13bcb @magic.fr. 1068

VOL. 42, 1998

RAT MODEL OF CONCURRENT INFECTIONS

1069

RESULTS

FIG. 1. Schematic representation of the experimental protocol for the tripleinfection model. IV, bacteria were inoculated intravenously; IP, tachyzoites of T. gondii were inoculated intraperitoneally; #, rats were autopsied within 10 h of death for determination of the pathogens.

weeks of corticosteroid immunosuppression, the rats were inoculated intraperitoneally with 1.5 3 107 tachyzoites of the virulent RH strain of T. gondii. Assessment of different infections. The different infections were assessed as follows. The numbers of P. carinii cysts in lung tissue were counted after enzymatic digestion and toluidine blue O staining of the lung tissue, and the T. gondii organisms in lung, brain, liver, and spleen tissue specimens were titrated by a tissue culture method and an indirect immunofluorescence assay as described previously (11). For M. avium, lung, spleen, and liver tissue specimens and blood specimens were stored at 280°C until they were processed. The tissues were homogenized in 1 ml of sterile distilled water with a glass homogenizer. Homogenates were decontaminated by a procedure which has no effect on the yield of M. avium and which is routinely used for the isolation of Mycobacterium tuberculosis (21). Serial 10-fold dilutions of homogenates were cultured on 7H11 agar plates supplemented with oleic acid, albumin, dextrose, and catalase enrichment (Difco Laboratories, Detroit, Mich.) (30). The colonies were counted after 14 days at 37°C, and the numbers of CFU per gram of tissue were calculated. Monitoring of development of infection in control rats. After the evaluation of the basal level of P. carinii infection in 8 rats, 45 rats were immunosuppressed to follow the development of (i) a simple P. carinii infection after 5 and 7 weeks of immunosuppression (10 rats), (ii) a dual infection with P. carinii and M. avium (20 rats), (iii) a dual infection with P. carinii and T. gondii (5 rats), and (iv) a triple infection with P. carinii, T. gondii, and M. avium (10 rats). Evaluation of drug activity in the triple-infection model. Forty rats were treated with combinations of drugs from the initiation of the corticosteroid treatment to the end of the experiment. Each drug was administered by the oral route 5 days a week for 5 weeks and then every day after T. gondii inoculation until death or killing of the animals after 7 weeks. The choice of drug combinations was guided by their efficacies in the treatment of P. carinii and T. gondii in the rat model of dual infection as shown previously, as follows: rifabutin at 100 mg/kg of body weight plus atovaquone at 100 mg/kg (8), roxithromycin at 200 mg/kg plus sulfamethoxazole at 20 mg/kg or dapsone at 50 mg/kg (9), PS-15 at 25 mg/kg plus dapsone at 25 mg/kg (10), and trimethoprim at 20 mg/kg plus sulfamethoxazole at 100 mg/kg (11). In addition, the synergistic activity of clarithromycin at 200 mg/kg plus a low dose of sulfamethoxazole (20 mg/kg) or plus atovaquone at 100 mg/kg was also examined because these combinations were shown to be active against P. carinii in a rat model (1) and against T. gondii in a mouse model (35). The goal from the use of these drug combinations was to extend their activities against M. avium infections: macrolides such as clarithromycin are one of the most effective drugs against the M. avium complex in humans (31, 36), roxithromycin has been shown to have efficacy against the M. avium complex in vitro and in vivo (5, 33, 40), and rifabutin has also been shown to have efficacy against M. avium complex infection in rat and mouse models (6, 26). Clarithromycin (Abbott Laboratories, North Chicago, Ill.), sulfamethoxazole (Sigma, Paris, France), roxithromycin (Roussel Uclaf, Romainville, France), dapsone (Rho ˆne-Poulenc-Rorer, Antony, France), rifabutin (Pharmacia Upjohn Laboratories, Milan, Italy), and PS-15 (Jacobus Pharmaceutical Co. Inc., Princeton, N.J.), all of which were in the powder form, were prepared in 1% carboxymethyl cellulose in sterile 0.9% saline solution and were briefly sonicated. Atovaquone (Wellcome Foundation, Beckenham, United Kingdom) was used in the suspension form, and trimethoprim combined with sulfamethoxazole (Roche, Neuilly-sur-Seine, France) was used in the pediatric solution form. Statistical analysis. P. carinii cyst counts, T. gondii burdens, and the numbers of M. avium CFU (per gram of tissue) were expressed as the mean log value 6 1 standard deviation. Results were analyzed by one-way analysis of variance, and each pair of groups of interest was compared by Bonferroni’s adjusted t test.

Monitoring of P. carinii, T. gondii, and M. avium infections in control rats. (i) Development of P. carinii in single, dual, and triple infections. The time course of P. carinii development as a single infection is presented in Table 1. The baseline value was log 3.3 6 0.6 cysts/g of lung, reflecting the latent infection. After 5 and 7 weeks of immunosuppression, the levels reached log 6.3 6 0.8 and log 7.2 6 0.3 cysts/g, respectively (P , 0.01). After 5 and 7 weeks of immunosuppression, the levels of P. carinii cysts in rats inoculated with M. avium complex reached log 6.1 6 0.5 and log 7.2 6 0.2 cysts/g, respectively (P , 0.01). After 5 weeks of immunosuppression the P. carinii cyst number was log 6.2 6 0.5/g in rats infected with T. gondii. When rats were infected with T. gondii and M. avium, the P. carinii cyst number was log 6.4 6 0.6/g. When these values were compared at the same time of immunosuppression, there was no difference between these values. Therefore, the P. carinii infection was not modified by the infection with the two other pathogens. (ii) Development of T. gondii in dual and triple infections. Rats latently infected with P. carinii and inoculated with T. gondii after 5 weeks of immunosuppression died by 6.2 6 1.8 days postinoculation; pleural fluid was found, and all examined organs exhibited T. gondii infection (Table 2). Of the 10 rats latently infected with P. carinii, inoculated with M. avium complex, and then challenged with T. gondii at week 5, 9 died by 5.7 6 0.2 days. The mean number of days to death, pleural fluid volumes, and T. gondii burdens in the organs were similar to those obtained for rats infected with P. carinii and T. gondii. Therefore, T. gondii infection was not modified by M. avium infection. (iii) Development of M. avium in dual and triple infections. Rats latently infected with P. carinii, inoculated with M. avium, and killed after 5 and 7 weeks of immunosuppression showed similar levels of disseminated M. avium complex infection in their lungs, spleens, livers, and blood (Table 3). Rats inoculated with the third pathogen (T. gondii) died 5.7 6 0.2 days after inoculation; bacteria were found in the pleural fluid, and the levels of M. avium complex infection in TABLE 1. P. carinii infection as a single infection, as a dual infection with T. gondii or M. avium, and as a triple infection with T. gondii and M. avium Infection and duration (wk) of immunosuppression

Simple P. carinii infection 0 5 7 P. carinii and M. avium infection 5 7 P. carinii and T. gondii infection until deathc P. carinii, T. gondii, and M. avium infection until deathd a

No. of rats

Outcome after T. gondii inoculation

8 5 5

3.3 6 0.6 6.3 6 0.8a 7.2 6 0.3a

10 10

6.1 6 0.5a 7.2 6 0.2a

5

Death

6.2 6 0.5a

9/10b

Death

6.4 6 0.6a

P , 0.01 compared to the value at week 0 (latent infection). One rat survived and was sacrificed at day 14. Rats died 6.2 6 1.8 days after T. gondii inoculation. d Rats died 5.7 6 0.2 days after T. gondii inoculation. b c

P. carinii count (log10) in lung

1070

BRUN-PASCAUD ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

TABLE 2. Characteristics of T. gondii infection in immunosuppressed rats as a dual infection with P. carinii and as a triple infection with P. carinii and M. avium Infection

No. of rats

Outcome after T. gondii inoculation (no. of days)a

Pleural fluid vol (ml)

Pleural fluid

Lung

Brain

Spleen

Liver

T. gondii and P. carinii infection T. gondii, P. carinii, and M. avium infection

5 9/10b

Death (6.2 6 1.8) Death (5.7 6 0.2)

4.8 6 0.9 3.4 6 1.8

2.2 6 1.5 3.6 6 1.5

4.7 6 0.6 4.7 6 1.9

3.1 6 0.9 3.7 6 0.6

4.8 6 0.5 4.7 6 1.0

6.6 6 0.4 6.1 6 0.9

a b

T. gondii count (log10)

T. gondii tachyzoites were inoculated intraperitoneally after 5 weeks of immunosuppression. One rat survived and was sacrificed at day 14.

their organs were similar to those observed in rats with dual P. carinii and M. avium infections. Therefore, M. avium infection in rats infected with P. carinii was not modified by T. gondii infection. Evaluation of drug activity in the triple-infection model. The results of the evaluation of drug activity are indicated in Table 4 for P. carinii and T. gondii infections and in Fig. 2A to D for M. avium complex infection. Untreated rats died 5.7 6 0.2 days after inoculation of T. gondii; 3.4 6 1.8 ml of pleural fluid was found. T. gondii burdens were found in the pleural fluid, lungs, brains, spleens, and livers, P. carinii cysts were found in the lungs, and M. avium was found in the pleural fluid, lungs, spleens, livers, and blood. Thirty-eight treated rats survived and were killed after 7 weeks of treatment. Two rats died during the course of treatment: one in the rifabutin-atovaquone group (shortly after gavage) and the other in the PS15–dapsone group (from a nonmycobacterial infection). The results indicate that (i) P. carinii infection was fully prevented in all treated rats, whereas it was not prevented in untreated control rats; (ii) T. gondii was found only in the spleen of one rat treated with the PS-15–dapsone combination; and (iii) compared to the levels of M. avium complex organisms obtained in untreated rats, the CFU counts after the administration of either clarithromycin-sulfamethoxazole, clarithromycin-atovaquone, rifabutin-atovaquone, roxithromycin-dapsone, or roxithromycin-sulfamethoxazole combinations were significantly decreased in the lungs, spleen, and liver (P , 0.01) (Fig. 2A, B, and C) and no bacteremia was detected (limit of detection, 10 CFU/mL) (Fig. 2D). The combinations PS-15– dapsone and trimethoprim-sulfamethoxazole were not found to be efficacious against M. avium complex infection, although a slight decrease in CFU counts in blood was observed. DISCUSSION In the setting of human immunodeficiency virus (HIV) infection, improved strategies for prophylaxis for the most frequent and severe opportunistic infections need to be de-

veloped. Although the advances in antiretroviral therapy have markedly decreased the level of progression of immune system deterioration and have improved the survival of patients with HIV infection, large numbers of HIV-infected patients remain at risk for the development of opportunistic infections. New challenges have emerged. While prophylaxis for pneumocystosis is well established, opportunistic infections associated with more severe immunosuppression such as localized cytomegalovirus infections or disseminated M. avium complex infection continue to occur (22). This phenomenon justifies the need to extend the spectra of prophylactic regimens by taking into account the activities of compounds alone or in combination against several pathogens. In addition, the introduction of antiHIV protease inhibitors which strongly interact with cytochrome P-450 have raised new pharmacological difficulties from the use of multiple drugs against opportunistic pathogens. Thus, we extended the model of a double infection (11) to a model of a triple infection that includes M. avium complex infection because of the frequency of occurrence of such infections in immunosuppressed patients and the susceptibility of M. avium to the drugs already used against T. gondii. In the present model, active infections with P. carinii, T. gondii, and M. avium were obtained. The levels of P. carinii burdens in the lungs were similar to those evaluated in the lungs of corticosteroid-treated rats, which is a model routinely used by numerous investigators (23) and which has been proved to be highly useful in predicting the efficacies of drugs in humans (24, 39). T. gondii infection in multiple organs was successfully achieved after inoculation of rats with the virulent RH strain, and parasite burdens were not different from those obtained in mouse models previously used to screen the activities of drugs against T. gondii (32). Although this model of subacute disseminated infection differs from the infection reactivation observed in immunosuppressed patients, it is nonetheless predictive of the clinical activities of anti-T. gondii drugs (11). In addition to T. gondii counts, the efficacies of drugs can be assessed by comparing the survival rates since toxoplasma infections are lethal. We also found that M. avium complex in-

TABLE 3. Characteristics of M. avium complex infection in immunosuppressed rats as a dual infection with P. carinii and as a triple infection with P. carinii and T. gondii Infection and duration (wk) of immunosuppression

M. avium and P. carinii infection 5 7 M. avium, P. carinii, and T. gondii infection until deatha a b c

No. of rats

M. avium complex count (log10) Pleural fluid

10 10 9/10b

Death occurred 5.7 6 0.2 days after T. gondii inoculation. One rat survived and was sacrificed at day 14. The determination could be done for only one rat.

1.5 6 0.6

Lung

Spleen

Liver

Blood

5.0 6 0.3 5.5 6 0.3

6.5 6 0.4 6.6 6 0.2

6.1 6 0.2 6.4 6 0.1

1.6 6 0.7 2.4 6 1.0

5.3 6 0.6

6.4 6 0.6

6.0 6 0.5

4c

VOL. 42, 1998

RAT MODEL OF CONCURRENT INFECTIONS

1071

TABLE 4. Activities of different drug regimens on P. carinii and T. gondii burdens in tissues in the rat model of concurrent pneumocystosis, toxoplasmosis, and M. avium complex infection Drugs (dose [mg/kg])a

None CLA (200) and SMZ (20) CLA (200) and ATO (100) RIFA (100) and ATO (100) ROX (200) and DAP (50) ROX (200) and SMZ (20) PS-15 (25) and DAP (25) TMP (20) and SMZ (100)

Outcome after T. gondii inoculationb

Death Sacrifice Sacrifice Sacrifice Sacrifice Sacrifice Sacrifice Sacrifice

Day

5.7 6 0.2 14 14 14 14 14 14 14

No. of rats c,d

9/10 10 5 5d,f 5 5 5d,g 5

Pleural fluid vol (ml)

3.4 6 1.8 0 0 0 0 0 0 0

Pleural fluid

Lung

Brain

Spleen

Liver

P. carinii count (log10) in the lung

3.6 6 1.5

4.7 6 1.9 0e 0e 0e 0e 0e 0e 0e

3.7 6 0.6 0e 0e 0e 0e 0e 0e 0e

4.7 6 1.0 0e 0e 0e 0e 0e 0.3 6 0.7e 0e

6.1 6 0.9 0e 0e 0e 0e 0e 0e 0e

6.4 6 0.6 2.5 6 0.5e 2.8 6 0.5e 2.8 6 0.9e 2.5 6 0.5e 2.4 6 0.5e 2.5 6 0.6e 2.7 6 0.7e

T. gondii count (log10)

a The drugs clarithromycin (CLA), sulfamethoxazole (SMZ), atovaquone (ATO), rifabutin (RIFA), roxithromycin (ROX), dapsone (DAP), PS-15, and trimethoprim (TMP) were given per os 5 days per week for 5 weeks and daily after T. gondii inoculation (at week 5) until death or sacrifice. b T. gondii tachyzoites were inoculated intraperitoneally after 5 weeks of immunosuppression. Sacrifice indicates that the rats were killed by the investigators. c One rat survived and was sacrificed at day 14. d The levels of infections detected were similar to those found after the other rats were killed and the values were pooled. e P , 0.01 compared to the values for untreated controls. f One rat died after a gavage at day 11. g One rat died from bacterial infection at day 13.

fection could be established in several organs and blood, although the level of infection was slightly lower than that in immunodeficient mouse models (19). The rationale for testing several combinations of drugs with potential activity against at least one pathogen was to validate this new model for the evaluation of the activities of drugs. The results confirm that several drug combinations are interesting candidates against the three infections and suggest the ability of this model to predict drug activity because the results are consistent with those obtained with classical one-pathogen models. The combination of a macrolide (clarithromycin or roxithromycin) with a sulfonamide or dapsone reduced the burdens of the three microorganisms and prevented mortality. The eradication of M. avium complex infection from tissues was not achieved, although bacteremia was abolished, which is consistent with observations with other animal models and for humans (6, 7, 13, 18, 26, 30). Clarithromycin plus atovaquone or rifabutin plus atovaquone was as effective as the previous combinations and could avoid the frequent undesirable sulfonamide- or dapsone-related side effects which are particularly common in HIV-infected patients (38). The anti-P. carinii and anti-T. gondii activities of the rifabutin-atovaquone combination are consistent with those described in recent studies demonstrating that the combination has synergistic activity against P. carinii (14) and T. gondii (3, 34). In contrast, trimethoprimsulfamethoxazole was not efficacious against M. avium complex infection in our study, and the results of our study do not support the clinical efficacy of that combination suggested by one group (16). PS-15 plus dapsone appeared to be effective only against P. carinii and T. gondii infections, confirming previous observations (10). No significant activity against the M. avium complex was observed, which is in contrast to the observations made in vitro by other teams (28, 37). The extrapolation of the data regarding the activities of the drug combinations to their clinical use in HIV-infected patients should be done with caution. It must be kept in mind that the mechanism of immunosuppression induced by corticosteroids is far different from the mechanism of HIV-related immunodeficiency and that the means of acquisition of each opportunistic infection in animals with corticosteroid-induced immunosuppression differ from those in patients with AIDS. Nevertheless, previous murine models of single opportunistic infections based on a similar experimental approach have been shown to be reliable for the preclinical evaluation of drug efficacy.

In conclusion, a rat model of triple infection with P. carinii, T. gondii, and M. avium has been designed and has been used to demonstrate the efficacies of several drug combinations. Compared to separate models of single infections, which often use different animal species, this model offers the opportunity to evaluate simultaneously the activities of drugs against different pathogens. The use of a single immunocompromised host is an advantage for the more rapid identification of candidate drug regimens with activity against opportunistic patho-

FIG. 2. Activities of different drug regimens on M. avium levels in lungs (A), spleens (B), livers (C), and blood (D) in the rat model of concurrent pneumocystosis, toxoplasmosis, and M. avium infection. p, P , 0.01 versus untreated controls. (a), value for one rat. See footnote a of Table 4 for definitions of abbreviations.

1072

BRUN-PASCAUD ET AL.

gens and for the evaluation of the activities of drugs under similar pharmacokinetic conditions. ACKNOWLEDGMENTS This study was supported in part by a grant from the Agence Nationale pour la Recherche sur le SIDA and by SIDACTION. REFERENCES 1. Adler, J., M. Mitten, N. Shipkowitz, L. Hernandez, Y. Hui, K. Marsh, and J. Clement. 1994. Treatment of experimental Pneumocystis carinii infection by combination of clarithromycin and sulphamethoxazole. J. Antimicrob. Chemother. 33:253–263. 2. Araujo, F. G., J. Huskinson, and J. S. Remington. 1991. Remarkable in vitro and in vivo activities of the hydroxynaphthoquinone 566C80 against tachyzoites and tissue cysts of Toxoplasma gondii. Antimicrob. Agents Chemother. 35:293–299. 3. Araujo, F. G., T. Slifer, and J. S. Remington. 1994. Rifabutin is active in murine models of toxoplasmosis. Antimicrob. Agents Chemother. 38:570–575. 4. Bartlett, M. S., J. A. Fishman, S. F. Queener, M. M. Durkin, M. A. Jay, and J. W. Smith. 1988. New rat model of Pneumocystis carinii infection. J. Clin. Microbiol. 26:1100–1102. 5. Bermudez, L. E., P. Kolonoski, and L. S. Young. 1996. Roxithromycin alone and in combination with either ethambutol or levofloxacin for disseminated Mycobacterium avium infections in beige mice. Antimicrob. Agents Chemother. 40:1033–1035. 6. Brown, S., F. Edwards, E. Bernar, W. Tong, and D. Armstrong. 1993. Azithromycin, rifabutin, and rifapentine for treatment and prophylaxis of Mycobacterium avium complex in rats treated with cyclosporine. Antimicrob. Agents Chemother. 37:398–402. 7. Brown, S. T., F. F. Edwards, E. M. Bernard, Y. Niki, and D. Armstrong. 1991. Progressive disseminated infection with Mycobacterium avium complex after intravenous and oral challenge in cyclosporine-treated rats. J. Infect. Dis. 164:922–927. 7a.Brun-Pascaud, M. G. Bertrand, F. Chau, P. Rajagopalan-Levasseur, L. Garry, F. Derouin, and P. M. Girard. 1996. Rat model of concurrent Pneumocystis carinii (Pc), Toxoplasma gondii (Tg) and Mycobacterium avium complex (MAC) infections for assessment of multiple prophylaxis, abstr. I96, p. 203. In Program and abstracts of the 36th Interscience Conference on Antimicrobial Agents and Chemotherapy. American Society for Microbiology, Washington, D.C. 8. Brun-Pascaud, M., F. Chau, F. Derouin, and P. M. Girard. 1996. Evaluation of the activities of rifabutin combined with atovaquone or low-dose of cotrimoxazole for prevention of pneumocystosis and toxoplasmosis in a dual infection rat model. J. Eukaryot. Microbiol. 43:14S–15S. 9. Brun-Pascaud, M., F. Chau, F. Derouin, and P. M. Girard. 1998. Experimental evaluation of roxithromycin combined with dapsone or sulfamethoxazole on pneumocystosis and toxoplasmosis dual infections in rat model. J. Antimicrob. Chemother. 41(Suppl. B):57–62. 10. Brun-Pascaud, M., F. Chau, L. Garry, D. Jacobus, F. Derouin, and P. M. Girard. 1996. Combination of PS-15, epiroprim, or pyrimethamine with dapsone in prophylaxis of Toxoplasma gondii and Pneumocystis carinii dual infection in a rat model. Antimicrob. Agents Chemother. 40:2067–2070. 11. Brun-Pascaud, M., F. Chau, A. M. Simonpoli, P. M. Girard, F. Derouin, and J. J. Pocidalo. 1994. Experimental evaluation of combined prophylaxis against murine pneumocystosis and toxoplasmosis. J. Infect. Dis. 170:653–658. 12. Carr, A., B. Tindall, B. J. Brew, D. J. Marriott, J. L. Harkness, R. Penny, and D. A. Cooper. 1992. Low-dose trimethoprim-sulfamethoxazole prophylaxis for toxoplasmosis encephalitis in patients with AIDS. Ann. Intern. Med. 117: 106–111. 13. Cohen, Y., C. Perronne, T. Lazard, C. Truffot-Pernot, J. Grosset, J. L. Vilde´, and J. J. Pocidalo. 1995. Use of normal C57BL/6 mice with established Mycobacterium avium infections as an alternative model for evaluation of antibiotic activity. Antimicrob. Agents Chemother. 39:735–738. 14. Comley, J. C. W., and A. M. Sterling. 1995. Effect of atovaquone and atovaquone drug combinations on prophylaxis of Pneumocystis carinii in SCID mice. Antimicrob. Agents Chemother. 39:806–811. 15. Derouin, F., C. Piketty, C. Chastang, F. Chau, B. Rouveix, and J. J. Pocidalo. 1991. Antitoxoplasma effects of dapsone alone and combined with pyrimethamine. Antimicrob. Agents Chemother. 35:252–255. 16. Fraser, I., I. Macintosh, and E. G. Wilkins. 1994. Prophylactic effect of co-trimoxazole for Mycobacterium avium complex infection: a previously unreported benefit. Clin. Infect. Dis. 19:211. 17. Frenkel, J. K., J. T. Good, and J. A. Schultz. 1966. Latent pneumocystis infection in rats, relapse and chemotherapy. Lab. Invest. 15:1559–1577. 18. Gangadharam, P. R. 1995. Beige mouse model for Mycobacterium avium complex disease. Antimicrob. Agents Chemother. 39:1647–1654. 19. Gangadharam, P. R. J., D. R. Ashtekar, D. L. Flasher, and N. Du ¨zgu ¨nes. 1995. Therapy of Mycobacterium avium complex infections in beige mice with streptomycin encapsulated in sterically stabilized liposomes. Antimicrob. Agents Chemother. 39:725–730.

ANTIMICROB. AGENTS CHEMOTHER. 20. Girard, P. M., R. Landman, C. Gaudebout, et al. 1993. A controlled trial of dapsone-pyrimethamine versus pentamidine aerosols for primary prophylaxis of Pneumocystis carinii pneumonia and opportunistic toxoplasmosis in patients with human immunodeficiency virus infection. N. Engl. J. Med. 328:1514–1520. 21. Grosset, J., C. Truffot, J. Fermanian, and H. Lecoeur. 1982. Activite´ ste´rilisante des diffe´rents antibiotiques dans la tuberculose expe ´rimentale de la souris. Pathol. Biol. 30:444–446. 22. Hoover, D. R., A. J. Saah, A. J. Bacellar, J. Phair, R. Detels, R. Anderson, R. A. Kaslow, for the Multicenter AIDS Cohort Study. 1993. Clinical manifestations of AIDS in the era of pneumocystis prophylaxis. N. Engl. J. Med. 329:1922–1926. 23. Hughes, W. T. 1989. Animal models for Pneumocystis carinii pneumonia. J. Protozool. 36:41–45. 24. Hughes, W. T., J. T. Killmar, and H. S. Oz. 1994. Relative potency of 10 drugs with anti-Pneumocystis carinii activity in animal model. J. Infect. Dis. 170:906–911. 25. Kaplan, J. E., H. Masur, K. K. Holmes, C. M. Wilfert, R. Sperling, S. A. Baker, C. Braun Trapnell, K. A. Freedberg, D. Cotton, W. G. Powderly, and H. W. Jaffe, and the USPHS/IDSA Prevention of Opportunistic Infections Working Group. 1995. USPHS/IDSA guidelines for the prevention of opportunistic infections in persons infected with human immunodeficiency virus: an overview. Clin. Infect. Dis. 21:S12–S31. 26. Lazard, T., C. Perronne, J. Grosset, J. L. Vilde´, and J. J. Pocidalo. 1993. Clarithromycin, minocycline, and rifabutin treatments before and after infection of C57BL/6 mice with Mycobacterium avium. Antimicrob. Agents Chemother. 37:1690–1692. 27. Masur, H. 1992. Prevention and treatment of Pneumocystis pneumonia. N. Engl. J. Med. 327:1853–1860. 28. Meyer, S. C. C., S. K. Majumder, and M. H. Cynamon. 1995. In vitro activities of PS-15, a new dihydrofolate reductase inhibitor, and its cyclic metabolite against Mycobacterium avium complex. Antimicrob. Agents Chemother. 39:1862–1863. 29. Ostroff, S. M., R. A. Spiegel, J. Feinberg, C. A. Benson, and C. R. Horsburgh. 1995. Preventing disseminated Mycobacterium avium complex disease in patients infected with human immunodeficiency virus. Clin. Infect. Dis. 21: 72S–76S. 30. Perronne, C., Y. Cohen, C. Truffot-Pernot, J. Grosset, J. L. Vilde´, and J. J. Pocidalo. 1992. Sparfloxacin, ethambutol, and cortisol receptor inhibitor RU-40 555 treatment for disseminated Mycobacterium avium complex infection of normal C57BL/6 mice. Antimicrob. Agents Chemother. 36:2408–2412. 31. Pierce, M., S. Crampton, D. Henry, L. Heifets, A. LaMarca, M. Montecalvo, G. Wormser, H. Jablonowski, J. Jemsek, M. Cynamon, B. Yangco, G. Notario, and C. Craft. 1996. A randomized trial of clarithromycin as prophylaxis against disseminated Mycobacterium avium complex infection in patients with advanced acquired immunodeficiency syndrome. N. Engl. J. Med. 335: 384–391. 32. Piketty, C., F. Derouin, B. Rouveix, and J. J. Pocidalo. 1990. In vivo assessment of antimicrobial agents against T. gondii. Quantification of parasites in blood, lungs and brain of infected mice. Antimicrob. Agents Chemother. 34: 1467–1472. 33. Rastogi, N., V. Labrousse, and A. Bryskier. 1995. Intracellular activities of roxithromycin used alone and in association with other drugs against Mycobacterium avium complex in human macrophages. Antimicrob. Agents Chemother. 39:976–978. 34. Romand, S., C. Della Bruna, R. Farinotti, and F. Derouin. 1996. In vitro and in vivo effect of rifabutin alone or combined with atovaquone against Toxoplasma gondii. Antimicrob. Agents Chemother. 40:2015–2020. 35. Romand, S., M. Pudney, and F. Derouin. 1993. In vitro and in vivo activities of the hydroxynaphthoquinone atovaquone alone or combined with pyrimethamine, sulfadiazine, clarithromycin, or minocycline against Toxoplasma gondii. Antimicrob. Agents Chemother. 37:2371–2378. 36. Shafran, S., J. Singer, D. Zarowny, P. Phillips, I. Salit, S. Walmsley, I. Fong, J. Gill, A. Rachlis, R. Lalonde, M. Fanning, C. Tsoukas, for the Canadian HIV Trials Network Protocol 010 Study Group. 1996. A comparison of two regimens for the treatment of Mycobacterium avium complex bacteremia in AIDS: rifabutin, ethambutol, and clarithromycin versus rifampin, ethambutol, clofazimine, and ciprofloxacin. N. Engl. J. Med. 335:377–383. 37. Shah, L., M. DeStephano, and M. Cynamon. 1996. Enhanced in vitro activity of WR99210 in combination with dapsone against Mycobacterium avium complex. Antimicrob. Agents Chemother. 40:2644–2645. 38. United States Public Health Task Force on Antipneumocystis Prophylaxis for Patients with Human Immunodeficiency Virus Infection. 1992. Recommendations for prophylaxis against Pneumocystis carinii pneumonia for adults and adolescents infected with human immunodeficiency virus. Morbid. Mortal. Weekly Rep. 41:1–11. 39. Walzer, P. D., J. Foy, P. Steele, and M. White. 1992. Treatment of experimental pneumocystosis: review of 7 years of experience and development of a new system for classifying antimicrobial drugs. Antimicrob. Agents Chemother. 36:1943–1950. 40. Young, L. S., L. E. Bermudez, M. Wu, and C. B. Inderlied. 1995. Potential role of roxithromycin against the Mycobacterium avium complex. Infection 23:S28–S32.